Reproducing SCOUTER

All code used for our (accepted) submission to the ML Reproducibility Challenge 2021, where we reproduced the experiments of the following paper:

Li, L., Wang, B., Verma, M., Nakashima, Y., Kawasaki, R., & Nagahara, H. (2021). SCOUTER: Slot attention-based classifier for explainable image recognition. In Proceedings of the IEEE/CVF International Conference on Computer Vision (pp. 1046-1055).

SCOUTER PDF & SCOUTER Code

Our paper can be found here

Downloading trained models

A single trained model for each experimental setup can be downloaded from here.

Usage

Examples of how to run the experiments can be found in produce_results.ipynb.

Cite

@article{maarten_burger_2022_6574641,
  author       = {Maarten Burger and
                  Kaya ter Burg and
                  Sam Titarsolej and
                  Selina Jasmin Khan},
  title        = {{[Re] Reproducibility Study - SCOUTER: Slot 
                   Attention-based Classifier for Explainable Image
                   Recognition}},
  journal      = {ReScience C},
  year         = 2022,
  volume       = 8,
  number       = 2,
  pages        = {#8},
  month        = may,
  doi          = {10.5281/zenodo.6574641},
  url          = {https://doi.org/10.5281/zenodo.6574641
}